Can crimes and crashes be blamed on bad genes?
On Friday, Nature reported details of an Italian case in which a court, for the first time in Europe, reduced a criminal sentence based on a genetic theory of behavior . According to the report:
Pietro Pietrini, a molecular neuroscientist at Italy's University of Pisa, and Giuseppe Sartori, a cognitive neuroscientist at the University of Padova, conducted a series of tests [on the defendant] and found abnormalities in brain-imaging scans and in five genes that have been linked to violent behaviour—including the gene encoding the neurotransmitter-metabolizing enzyme monoamine oxidase A (MAOA). A 2002 study ... had found low levels of MAOA expression to be associated with aggressiveness and criminal conduct of young boys raised in abusive environments. In the report, Pietrini and Sartori concluded that [the defendant's] genes would make him more prone to behaving violently if provoked. ... On the basis of the genetic tests, Judge Reinotti docked a further year off the defendant's sentence, arguing that the defendant's genes "would make him particularly aggressive in stressful situations". Giving his verdict, Reinotti said he had found the MAOA evidence particularly compelling.
Scientists interviewed by Nature say this kind of genetic determinism is too crude and tenuous to justify sentence reductions. Genes operate in relation to each other, not independently, and they produce different effects depending on environmental factors and population differences such as ethnicity. My colleague Dan Engber, who alerted me to the Nature report, has made similar points about brain scans and neuro-determinism . I'm highly sympathetic to his indictment of the field.
Despite these criticisms and the skepticism of judges, Nature reports that according to a database maintained by Nita Farahany of Vanderbilt law school,
in the past five years there have been at least 200 [U.S.] cases where lawyers have attempted to use genetic evidence to support the idea their clients' were predisposed to violent behaviour, depression or drug or alcohol abuse. In Britain, there have been at least 20 such cases in the past five years.
And don't be surprised if the next target of genetic determinists is car crashes. Neuroscientists from the University of California at Irvine have just laid out the case in a paper in Cerebral Cortex , accompanied by a catchy press release. The release is titled, " Bad driving may have genetic basis , UCI study finds." It reports:
People with a particular gene variant performed more than 20 percent worse on a driving test than people without it—and a follow-up test a few days later yielded similar results. About 30 percent of Americans have the variant. ... The driving test was taken by 29 people-22 without the gene variant and seven with it. They were asked to drive 15 laps on a simulator that required them to learn the nuances of a track programmed to have difficult curves and turns. Researchers recorded how well they stayed on the course over time. Four days later, the test was repeated. Results showed that people with the variant did worse on both tests than the other participants, and they remembered less the second time.
The release explains how the gene works:
This gene variant limits the availability of a protein called brain-derived neurotrophic factor during activity. BDNF keeps memory strong by supporting communication among brain cells and keeping them functioning optimally. When a person is engaged in a particular task, BDNF is secreted in the brain area connected with that activity to help the body respond.
The study, funded by the National Institutes of Health, will surely become fodder for defense attorneys in civil and criminal cases involving crashes or dangerous driving. And these lawyers won't be alone in their interest. In the press release, Steven Cramer, the neuroscientist who led the study, says, "I'd be curious to know the genetics of people who get into car crashes. I wonder if the accident rate is higher for drivers with the variant."
I bet every auto insurer would like to know the same thing.
I'm counting on Engber to pick apart the study and the press release. Let's hope the authors are oversimplifying the impact of genes on behavior. Because if the link between genes and driving performance is solid enough to justify reduced sentences and damage awards, it's hard to see why insurances shouldn't be allowed to test and charge you accordingly.